cloudflare-workers

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.80). This skill's SKILL.md includes code that fetches and processes arbitrary external content (e.g., the "Image Proxy" example uses fetch(imageUrl) from a request parameter and multiple examples/HTMLRewriter call fetch(request) to ingest origin responses), which clearly ingests untrusted, user-provided/public web content that the worker reads and acts on.